Page 25 - BH-2-3

Brain & Heart Digital tools for stroke and bleeding risk in AF

stroke and bleeding. It is responsible for 10% of strokes and The identification of stroke risk factors in patients
is associated with a two-fold increased risk of death. In with AF spurred the development of scoring systems
4
most patients with AF, the rate of thromboembolic events for risk estimation. CHADS , introduced in 2001,
2
is five- to eight-fold higher than bleeding rates. However, was the first widely adopted scoring system for this purpose.
5
patients on anticoagulation therapy face an increased risk The CHADS score was constructed by amalgamating the
2
of major bleeding, ranging from 1.4% to 3.4% while using independent risk factors identified from SPAF and AFI,
warfarin. A central issue in the management of AF is and its validation utilized the National Registry of AF
6,7
effectively preventing cardioembolic strokes while carefully (NRAF), a dataset assembled from Medicare claims of
balancing the bleeding risks associated with therapy. patients hospitalized for non-rheumatic AF who had
17
The selection of stroke risk reduction therapy is guided not received anticoagulation therapy. Congestive HF,
by the patient’s risk of stroke, bleeding risks associated hypertension, age ≥75 years, and diabetes mellitus were
each assigned 1 point in the score, whereas history of
with the therapy, and individual preferences. Given the
dynamic nature of these factors, it is crucial to periodically stroke or TIA was each assigned 2 points. In contrast to
reassess all decisions related to stroke prevention therapy subsequent scores, hypertension was defined as “any
history of hypertension” rather than using a numeric BP
and to make use of validated scores for risk estimation. threshold, and congestive HF (CHF) was defined as “recent
To support ongoing discussions and provide guidance to
both patients and providers, various digital tools have been heart failure exacerbation.” Scores were initially categorized
17
developed to augment the role of these risk scores. In this as low (0 – 1), moderate (2 – 3), and high (≥4) risk levels.
Subsequently, moderate risk was denoted as 1 point and
paper, we review the historical development of scores used high risk as ≥2 points. . During this time, treatment with
18
to estimate stroke and bleeding risks in AF, as well as the
current landscape of digital tools available for this purpose. warfarin, the only pharmacologic option available for
anticoagulation, was generally recommended for patients
2. Scores for estimating stroke risk in AF with a CHADS score of ≥2, irrespective of gender. 18
2
Over the years, at least 19 risk scores and 76 updates to Since the development of CHADS , efforts have been
2
risk scores have been published to estimate stroke risk in underway to refine scores for stroke risk assessment in
AF. Major scores for estimating stroke risk in AF include AF, particularly because a substantial number of patients
8
CHADS 2001), Framingham (2003), CHA DS -VASc with AF were placed into the moderate risk category,
2
2
2 (
(2010), ATRIA (2013), ABC (2016), GARFIELD-AF according to CHADS . An alternate scoring system based
2
(2017), and IMRS-VASc (2019) (Table 1). By the late 1980s, on Framingham Heart Study data was introduced in
19
the association between AF and stroke became apparent, 2003. Similar to the CHADS , advancing age, increasing
2
even in patients without mitral valve disease. The systolic blood pressure, female sex, diabetes, and history
9
Framingham Heart Study documented a dramatic increase of stroke or TIA were identified as stroke risk factors in
in AF incidence with each successive decade of age. In the Framingham score. However, the Framingham score
10
addition to age, independent risk factors for AF included assigns more points based on age and severity of risk factors.
diabetes, hypertension, valve disease, and congestive heart While the Framingham score improved the identification
failure (HF). The Stroke Prevention in AF (SPAF) I, II, of patients with AF at low risk for stroke and who may not
11
and III trials compared the efficacy of warfarin, aspirin, and have derived significant benefit from anticoagulation, it did
placebo in stroke prevention for patients with non-valvular not estimate risk reduction with anticoagulation therapy. In
AF. Collectively, SPAF I, II, and III demonstrated greater 2004, a study involving patients with non-valvular AF who
12
stroke reduction in patients prescribed with warfarin (with were taking aspirin revealed that CHADS had the greatest
2
a target international normalized ratio [INR] of 2.0 – 3.0) discrimination in identifying high-risk patients compared
compared to those on combination low-dose warfarin/ to the AFI, SPAF, and Framingham risk scores. However,
aspirin or aspirin monotherapy, and this established these scoring systems showed minimal differences in their
20
warfarin as the standard therapy for stroke prevention in ability to discriminate among low-risk patients.
AF. 13-15 The SPAF trials identified age, female sex, systolic In 2008, a validation study of the CHADS score revealed
2
blood pressure (BP)>160 mmHg, and prior stroke or that although it was a good predictor of stroke risk in
transient ischemic attack (TIA) as risk factors for stroke in patients with AF, incorporating additional factors such as
AF. At the same time, the AF investigators (AFI) pooled sex, extending age categories, and reweighing existing risk
12
data from five trials and identified age, hypertension, prior factors could result in improved accuracy. Addressing
21
stroke, or TIA, and diabetes mellitus as additional risk this, the Birmingham 2009 score, more popularly termed
factors for stroke in AF. 16 the CHA DS -VASc score, was introduced in 2010. The
22
2
2
Volume 2 Issue 3 (2024) 2 doi: 10.36922/bh.3068